A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 3, 19:00 UTC
[AMD] skip machete tests for rocm (#42326)
Signed-off-by: Hissu Hyvarinen <hissu.hyvarinen@amd.com>
hissu-hyvarinen committed
May 13, 2026 at 15:11 UTC
0a62f5eec967bd816be144ccf365e10c99c668db
1 file changed
+1
-1
tests/quantization/test_cutlass_w4a16.py
+1
-1
@@ -14,7 +14,7 @@ import torch
14
15
from vllm.platforms import current_platform
16
17
-if not current_platform.has_device_capability(90):
17
+if not current_platform.has_device_capability(90) or current_platform.is_rocm():
18
pytest.skip(
19
"Machete W4A16 requires Hopper (sm_90).",
20
allow_module_level=True,